The Nikon D3 DSLR


Nikon D3 DSLR

Introduction

Welcome to my thoughts and feelings about the new Nikon flagship, the D3. As with all of my reviews this will not contain much technical jargon or specifications for the camera as they can be found elsewhere. Instead it will concentrate on how the camera actually performs in the field and whether it does what is says on the tin. Specifically for this camera I will concentrate on exposure, autofocus, image quality, high ISO performance and some of the new features included on the D3. It will also be a review in two stages, a First Look which is the results of my first day with the camera and then a more in-depth look at everything after a few days of working with it.

The Nikon D3 has promised much and is the DSLR which Nikon photographers have been waiting for with a full-frame sensor, fully customisable options and a brand new AF system. So let’s see how it measured up.

First Look

Nikon kindly sent me one of the new 70-200 f2.8G VR lenses to use with the D3 which is great except that this is a much shorter lens than I would use at this time of the year for wildlife. I know that Nikon Pro Services are particularly busy now and all longer lenses are out and about so will have to make do with this lens for now, Wex Photographic have supplied me with a 1.7x teleconverter to help a little.

Picture info – all the pictures shown here are displayed as shot from the D3 and no colour manipulation has taken place. The D3 was set to a Picture Control of neutral, lossless NEF, colour space Adobe 1998 and no sharpening has been applied anywhere. The images were all processed using Adobe Lightroom 1.3.1 and again no colour corrections have been made. I just needed to get that off my chest!

My first shots with the D3 were taken this evening. I charged the battery, spent 30 minutes reading the manual and set it up for an autofocus test. Not just any autofocus test but an extreme one....fast running dogs in very low light. And I mean fast dogs, my two are speed demons and we took them up to the local field at sunset. I set up the D3 to have Auto ISO to get a shutter speed of 1/2500th – this is a great new feature where the D3 will find the correct ISO value to give me the shutter speed that I want. I can see many situations where this will be very useful. In this situation it set the ISO at 1400, which is much higher than any Nikon photographer would have dared to go before!

Then we unleashed the hounds and did several runs straight at the camera after the sun had set, like I said an extreme test. I should say that I had intended to do a test in the light but clouds intervened and that was the day over! I won't say that every shot from the sequences is in focus, there is no way that it can be with the low light and the forward speed of the hounds, but enough were razor sharp to warrant this a complete success. And the images that were sharp were RAZOR sharp...and at ISO 1400. I could not see much discernable noise in the background and the image detail was incredible, actually better than the 1D MK3 I have been using at high ISO. Check out the originals and crops below:



And then as the light got worse the ISO increased to 2000, so I naturally took some portraits! Here is the original:

And the aggressive crop into the eyes:

At ISO 2000 I have never seen so much detail in an image and remember I have not processed these or sharpened them in anyway. I have to say that I am very impressed by the D3 performance, both in terms of its AF system and low light high ISO performance and can’t wait to test it further…..if only I had a longer lens!

Well I have now had a few days to shoot with the D3 and now have a much better idea of what it can do. Some areas are still to be tested and I will attend to that this week and update before XMAS, but here are my findings so far. Rather than just test the D3 anywhere I have stuck to my specialisation and concentrated on wildlife and scenics. I think that it is important when judging a camera that you stick to something you know and have shot before, this gives it a fair test and allows me to form an opinion of how it performs. I took the D3 on two shoots, one was for roosting birds and the second was to a seal colony. The latter was designed to test the D3’s ability to work in extreme conditions as it was wet, sandy, windy and damn cold; it would also test the VR capability of the lens as my hands were freezing and shaking quite a bit. Anyway here goes, let’s start with exposure:

Exposure

I have left the D3 almost entirely on matrix metering and shot in a variety of situations including pointing at the sky, black subjects, black reflective subjects and white subjects so quite a varied test. In general I have found the matrix metering to be very accurate and I have only had to intervene a few times, usually to underexpose the images. In fact this has been a common thread throughout the exposures as most are on the bright side. When I look at the histogram it shows a great spread though from left to right in most cases, indicating a good tonal range. The peaks do seem to be towards the right and I get the impression that the metering may be slightly biased towards bright images in order to maximise the sensor performance and reduce any noise. In fact some of the histograms sit right against the RH axis which would normally mean overexposure and destination trash, but a single click in LightRoom and the histogram shifts to the left and they look great; so what looks like highlight burnout isn’t as I first thought.

I have dialed in a – 0.3 compensation to balance the exposures up slightly and now, after shooting about 1500 exposures with the D3 I can say that I have not deleted a single exposure because it is too dark or washed out. Plenty that I have messed up for various other reasons but none for exposure. Since I prefer the shoot right exposure technique anyway most of the images need a 2 second tweak under LightRoom to bring them back, but then they look punchy and alive. Of course I will continue to test the exposure meter over the coming days and update any further findings at the end but so far I think that it is pretty trustworthy. You can see some of the exposures in the rest of this review but here is something that the D3 produced that I did not think it would:

Not only did it lock onto the seals head but the exposure was spot on and the sun was not burnt out. A really amazing image for the exposure and so nicely balanced that I can send it straight out to clients without any post processing.

Image Area

By default the D3 produces a NEF that is 4256 x 2832 pixels which equates to an 8-bit TIF that is approximately 35 MB in size. When a DX lens is attached the D3 automatically crops the image to 2784 x 1848, which is approximately a 14.7 MB file which is about 20% smaller than A4. I found out about the DX crop when I attached my 17-55mm lens and the viewfinder changed to have a greyed out area around the edges. It was alarming at first but now I have found an interesting usage for it. When shooting the seals I used the 70-200 VR with the 1.7 converter which gave me an effective focal length of 340mm. For the seal pups this was no problem but trying to get the adults fighting in the sea I needed something beyond 500mm. Of course I could crop the original on the workstation but the D3 gives me another option. Using the FN button and the command dial I can switch from full-frame FX to the DX crop (or even 5:4 ratio) which takes my 340 up to 510mm.

 

These two images show the difference (yes I know they are hardly award winning but they make the point). I found this incredibly useful yesterday and can see times in the future when I just want to get a closer shot in camera. Of course the file is quite small, but it is good enough quality to allow me to stretch it upto 30MB without much noticeable reduction in quality. A useful idea and something that I thought you should know.

Shootability

Yes I know a new word for the dictionary but I wanted to test the D3 to see if it were “rouseworthy”. I demand a lot of cameras, they have to work in extreme environments and be able to shoot quickly and accurately. They also have to be straightforward and ergonomically simple to use. I have called all this shootability and over the past few days have really began to get to grips with the D3, anyway here are some findings:

  • Drive – a respectable 9 fps when shooting at the full image size (see below) which expands to 11 fps when the cropped size is used. The maximum values for both of these sizes can be changed in the menus.
  • Controls – this has been the biggest learning curve for me and I am still not there yet but I am getting a lot quicker. It is not quite second nature yet which dial to choose for what but I am finding that the rear multi-selector is well weighted and simple to use. My main gripe here is that when shooting in vertical mode the multi-selector dial is reachable but you have to stretch your fingers a little so I would like to suggest to our friends at Nikon to make it larger or put it in a better position for vertical shooting. Other than that it is all obvious.
  • Image Area select – I have covered this function previously but the default location for this is to press the FN button and the command dial. To get to the FN button, which is located on the front of the camera, requires some dexterity and I found that I had to take my eye from the viewfinder to do it. This means missed shots. So I suggest that you re-assign the Image Area select away from the FN (suggest using the virtual horizon function here) button and re-assign it using the menus to the AE-L / AF-L button on the back where it is much easier to get at. The D3 offers a high level of customisation and it is just a menu option to do this (f6).
  • Dual CF card slot – give them a medal, two CF cards can be used side by side which means I can load the camera with 2 x 8 GB cards. Awesome. Options are provided too to control how the two CF slots are used, i.e. when one fills up switch automatically to the other (it does this seamlessly), create a mirror image of the first on the second or write a JPEG to the second each time a NEF is written to the first. All useful options that I can see reasons for, I use the first by the way.
  • LCD – the screen is the largest on any Nikon dslr so far and has a beautiful colour balance that really looks like what I am seeing with my eyes.
  • Battery – is there one inside as I keep forgetting? I shot 1200 images yesterday in the cold with the VR switch on the lens on and the autofocus working overtime and the strength meter has not changed from full.

Image Quality

Of course the bottom line of most DSLRs comes down to quality, although I feel that the D3’s shootability (see below) is up there with it for photographers. I have been shooting the D3 in lossless compressed NEF and have not tested out the JPEGs at all so apologies to JPEG shooters out there. For you Nikon have put in a set of sophisticated picture controls so that you can control how the JPEG looks when it comes out of the camera, but I have not tested these so cannot comment.

So what can I say about the quality of the NEFs? Well so far it has been absolutely great and that is from someone who always demands the ultimate in quality from his cameras. I have shot the camera from 200 to 2000 and all of the images are usable; in fact I am now shooting with the D3 at ISO 400 all the time as the quality is great, there is little noise to speak of and I am getting less wasted shots due to low shutter speeds. Here are some test images for you to make your own judgements, please remember that all images are processed straight from the camera (i.e. they have not been colour corrected in photoshop), unsharpened and compressed for the web. Trust me they are sharp.

 

This is a test in flat light, which has always been a problem with all DSLRs as the images tend to be noisy and struggle for detail. As you can see the tight crop is full of detail and there is no visible noise in the shadows, so the D3 produced a usable image and it did so consistently in the grim light.

 

The light slightly improved to bright diffuse which is my favourite light for shooting. So here is the main image and again a cropped in version, the quality is awesome, especially when you consider that this is at ISO 400. Note the exposure too, a nice range of tones between the dark of the seal and the bright sky and waves behind, and a lot of nice detail in the darker areas of the face.

 

Beautiful isn’t it? Nothing more innocent and relaxing to spend time with than a week old Grey Seal pup in my books. Great quality, this time with an ISO of 250, I focused on the eyes and used an aperture of 5.6 to keep the shutter speed up as the light was starting to fade. Lovely quality on the face and the hairs really stand out even without sharpening, with just a touch of USM they jump from the page.

Most photographers shoot at ISO 200 in the belief that we get the best quality, I am one of those too and always try to shoot even lower at ISO 100 to get a better image. The Canon 1D MKIII convinced me that I could start to shoot at a higher ISO and now the D3 is pushing this boundary up further so that I will shoot now at ISO 400 by default. This means a higher average shutter speed and less throwaways due to camera shake.

High ISO Performance

The images that started this review certainly wetted my appetite to try the high ISO performance of the D3.

 

 

I managed to crawl close to one friendly bull who let me rattle off a sequence with different ISO’s, so in order here is ISO 1000, 2500 and 4000. Now remember that this is not in sunlight and not even diffuse light, it was overcast. The 2500 and 1000 images are barely discernable from one another and have excellent detail and well acceptable noise levels. But it is the ISO 4000 image that really surprised me. OK so there is a bit of noise in the dark areas but this looks like film grain and not the usual colour noise associated with high ISO images.

This image shows the kind of usage that I would have for this high ISO function. Taken at ISO 1000 when the light was down I did not have a tripod with me but wanted to record the scene as it was beautiful. So I set the ISO to 1000 and could handhold at a 1/400th second and get the image that you see here. High ISO images are for situations where you have to get the shot and there is no light with which to work and certainly Nikon have achieved that.

Autofocus

The autofocus system of the D3 has been completely revamped and this is one area that I have been really keen to test. All of the 1500 shots taken during this test have been taken using AF (as opposed to manual focus) and I have found it quick and responsive. So far I have had limited chance for some real action but later this week I will be putting it through the flying-falcon-like-a-rocket test so that should provide some more feedback in this area.

Existing Nikon users will be familiar of course with the three autofocus modes of Single Servo, Continuous Servo or manual. What is new though is the switch on the rear which controls how the individual AF points can be selected, you can either do it yourself completely (single-point AF), let the D3 help out when needed (Dynamic area-AF) or let the D3 sort it all out itself (Auto-area AF). Single-point AF is self explanatory and to select the point you want you use the rear multi-selector which has a helpful central button to automatically select the central point. Dynamic is similar, except that if you have continuous focus selected and the subject moves away from your autofocus point then the D3 will attempt to re-focus using information from the surrounding points. Finally Auto-area AF puts the D3 in complete control, it determines the focus points used and according to Nikon with type G or D lenses can distinguish human subjects from the background and select the correct focusing point.

The D3 allows you to customize how many AF points can be used from 9,21 and 51; remember that it is not always the biggest that is best. If you select 51 and Auto-area AF then the D3 gives an extra option called 3D tracking which uses the colour of the subject to keep focus on it when it moves across the frame.

It all sounds great, the question is does it work? Well so far I have had limited testing as I have just been using the 70-200 VR but can say that for normal AF work the D3 autofocus is highly accurate. By normal I mean something like a seal sitting there looking pretty and focusing on the eyes using a single point in Dynamic-Area AF and single-servo; I found the resulting images to be pin sharp across the board. The points were easy to select using the multi-selector but occasionally it seemed to develop a mind of its own and flick to another point after I had made my selection. I think that this is caused by it being very sensitive to the touch as this always happened when I moved my hand away; it is easily corrected but worth a mention here. Anyway here are some images:

Just to show that the autofocus works with difficult subjects away from the central point, I deliberately put these small in the frame and off-centre to see if I could throw the autofocus off. It was an accurate as I would expect from my Canon system.

 

Two shots showing the autofocus on continuous-servo Dynamic AF with a difficult background behind. It kept the lock well as the action changed around the frame and only once did it drift off to the background. My only regret is that I had not seen the high ISO shots beforehand else I would have shot this at ISO 800 and got enough shutter speed to freeze it in motion!

I had the D3 in full control here, 51 focusing points, continuous servo and Auto-area AF with dynamic 3D. The knot swirled in the air and I captured it as best I could, all 7 in the sequence are sharp but this has the nicest pattern. Why do I include this image the techno heads amoungst you might ask? Simply because all I can ask of any camera is to capture what I see and to react instantly to situations, with this image the D3 did just that.

Story so far

That has covered the main points of the D3 I think that I have found so far and I wanted to get something online at the start of this week. Hopefully I will be using a 200-400mm lens this week and so will get the chance to see how the D3 performs with a longer lens, I will also be testing the autofocus a lot more with fast birds in flight. I am not going to post any comments on the D3 yet except to say that we need to forget about the inevitable and very boring Canon vs. Nikon argument that is starting to rage. Look, the D3 is a wonderful camera and is without doubt (in my opinion) the best DSLR that Nikon have ever produced and quite possibly the best on the market today. I am saying this as someone who owns 30K of Canon gear and has never rated Nikon cameras before as being equal to what I use. Well those days are well and truly over and it has been a real pleasure to use the D3 and I am looking forward to further tests this week. Until then...

Update 17-12-07

Monday was a bit of a rush to get the review online and in doing so I forgot to mention several features that I have either tested or am in the process of testing:

  • Autofocus options – the autofocus system for the D3 is very customisable via a series of menus. Options are explained in plain English and the most useful for me controls whether you allow the D3 to shoot if autofocus is not attained. I am currently setting this to FOCUS as I want the D3 to only shoot when the subject is in focus but for certain situations I might choose to shoot with the opposite of this and the D3 gives me that option. I have added it to my favourites menu as it is something that I will need to change in a hurry.
  • Vertical shooting – in the main review I mentioned that it was more difficult to select focusing points when holding the camera vertically. Well I still stand by that but in the interests of being fair to Nikon I should say that I have found an option which attempts to cater for this. Custom setting f4 allows you to specify that you use the FUNC button and the command dials for focus point selection when in portrait mode. I have tried this out and think that it is actually more difficult to do it than using the multi-selector as the FUNC button is small and located on the front of the body, which makes it quite difficult to find in a hurry. So I still would like to see either a bigger multi-selector or one that is in a more balanced position for horizontal and vertical shooting.
  • 14-bit mode –There is an option to shoot NEFs in 14-bit mode rather than 12. This has the advantage of course of recording more colour tones in the scene but will produce a slightly larger file; personally this is not a trade off at all as I would rather get all the colour information that I can so I have been shooting in this the whole time. So far I can see no other downside and the camera appears to still work at 9 fps with the same buffer performance. Hopefully I will test this more in the coming days.
  • Auto ISO – something familiar to D2 users, Auto ISO allows you to set an aperture or shutter speed that you want and the D3 will set the ISO automatically needed to achieve it. In the past this could result in ISO settings of 2000 or more in low light which would render quite noisy images, with the D3 this is not a problem as I have shown above. I suspect that this will be a very useful tool to have.
  • Virtual horizon – sounds great and something which I really need as I have a reputation for drunken horizons, trust me I will be testing this!

That is all for now, hopefully more updates over the Xmas period as I will be shooting in the next few days.

Update 03-01-08

My final week of testing was curtailed by a combination of the weather and other factors so I only had the chance to test it with a captive owl shoot that I was doing. The day was misty and usually I would called it off because the shutter speeds would have been too low, but with the D3’s high ISO performance I decided to go ahead. The conditions were actually atmospheric and really inspiring to shoot in and the D3 allowed me to shoot at ISO 2000 and 4000 and get usable results. In fact I really like the look of these high ISO images and at ISO 4000 in these conditions the images looked like scanned transparencies; the noise in the background was consistent and subtle enough that I could live with it for any repro usages of the images. One factor that improved my shoot was the arrival of the 200-400 lens on loan, which gave me so much versatility in how to shoot on the day.

The quality at high ISO was amazing though; here is an image at ISO 1600 that is punchy, bright and alive.

Nikon D3 Sample Image

One area that I really wanted to test with the 200-400 was the AF capabilities of the D3 but the weather really put paid to it. Right at the end of the day though in the fading light I did manage to get a couple of decent flights from the Tawny Owl and tested the AF to a small degree. Now a Tawny Owl flies very erratically and takes off very fast, so the D3 had to react quickly.

Nikon D3 Sample Image

The results were pretty good, with 4 razor images out of 8, which is about the ratio that I would expect with my normal gear. Of course this does not answer the real questions about the D3’s AF system but I know a fair few sports pros that say it is awesome and I trust their opinion.

Conclusions

This has been a long review and one that I wanted to do so that I could be fair to all the Nikon clients of wex photographic. I have put the D3 through its paces in many different situations and each time it has more than been upto the task. The D3 has amazing ISO performance, produces images with a beautiful dynamic range and has the best AF system seen in a Nikon DSLR so far. So is the D3 worth upgrading to from a previous Nikon DSLR……absolutely 100%. Is it worth changing systems for from Canon? Well that must be your personal choice of course but already several well known pros have made the jump and I expect more to do the same soon, love us or hate us professionals live by the quality of our images so we must know something!

Copyright

This review is the copyright of Andy Rouse Wildlife Photography Ltd and Wex Photographic. No part of this review may be emailed, quoted or in any other way reproduced without the written permission of Andy Rouse. All opinions expressed are those of Andy Rouse and no third party has been consulted.

Back to top